Huo Zhenhua, Li Xiaoli, Chen Jing, et al. CMA global ensemble prediction using singular vectors from background field. J Appl Meteor Sci, 2022, 33(6): 655-667. DOI:  10.11898/1001-7313.20220602.
Citation: Huo Zhenhua, Li Xiaoli, Chen Jing, et al. CMA global ensemble prediction using singular vectors from background field. J Appl Meteor Sci, 2022, 33(6): 655-667. DOI:  10.11898/1001-7313.20220602.

CMA Global Ensemble Prediction Using Singular Vectors from Background Field

DOI: 10.11898/1001-7313.20220602
  • Received Date: 2022-05-30
  • Rev Recd Date: 2022-07-29
  • Available Online: 2022-11-21
  • Publish Date: 2022-11-17
  • China Meteorological Administration Global Ensemble Prediction System (CMA-GEPS) adopts singular vector method to generate initial perturbations. CMA-GEPS currently uses the initial analysis field from CMA Global Forecast System (CMA-GFS) data assimilation to calculate singular vector (ANSV). With this dependency, in the operational running procedures of CMA numerical weather prediction systems, the singular vector calculation starts when CMA-GFS analysis job is finished. With the improvement of model resolution, especially the horizontal resolution, the computation time of data assimilation analysis and the ensemble forecasts would be lengthened. Given relatively limited high-performance computational resources, it would bring great challenge for delivering the ensemble forecast products on time. ECMWF uses the data assimilation background field to calculate singular vector (FCSV), which could be implemented earlier than the computation of ANSV in the operational flow and then optimize the computation time for ensemble prediction system (EPS), and it shows that the performance of FCSV ensemble is comparable to ANSV ensemble.Based on CMA-GFS background field and SV calculation module, the feasibility of applying FCSV in CMA-GEPS is investigated. First, the spatial structures of ANSV and FCSV and their similarity index are analyzed. And then, two ensembles based on ANSV and FCSV are conducted for 10 cases in summer and autumn. The forecasts from ANSV ensemble and FCSV ensemble are comprehensively evaluated in terms of the ensemble prediction skill of barometric surface variables, the probability prediction of 24 hours accumulated precipitation in China, tropical cyclone track ensemble prediction skill, and the forecast skill of the minimum sea level pressure at tropical cyclone center. The results show that for the dominant extra-tropical singular vector in CMA-GEPS, the ANSV and FCSV have similar horizontal and vertical structures, their general similarity index is 0.6-0.8, and two ensembles have the comparable forecast skill over extratropics. For tropical singular vector which are only calculated when tropical cyclones are observed, their similarity index between ANSV and FCSV is relatively lower than that in extratropics, and FCSV ensemble shows slightly smaller ensemble spread but comparable error for tropical cyclone tracks. For the precipitation forecast, two ensembles have similar forecast skills for moderate to heavy rain. For mean sea level pressure forecast of strong tropical cyclone case, two ensemble have members showing the skill in terms of structures and magnitude. Therefore, it is feasible to apply FCSV in CMA-GEPS, and it could be an option to construct singular vector-based initial perturbations for future high-resolution operational CMA-GEPS.
  • Fig. 1  Potential temperature perturbation component(the shaded, unit:K) of the first singular vector, the second singular vector and the fifth singular vector(multiplied by 500) at 28 model level and the initial 500 hPa geopotential height(the contour, unit:gpm) in the Northern Hemisphere used to compute singular vectors corresponding to ANSV experiment and FCSV experiment with initial time of 1200 UTC 7 Sep 2020

    Fig. 2  Vertical structures of the potential temperature perturbation component(unit:K) of the first singular vector, the second singular vector and the fifth singular vector(multiplied by 500) at 50°N in the Northern Hemisphere corresponding to ANSV experiment and FCSV experiment with initial time of 1200 UTC 7 Sep 2020

    Fig. 3  Evolutions of root mean square error of the control forecast, root mean square error of ensemble mean and ensemble spread isobaric variables in the Northern Hemisphere

    Fig. 4  Evolutions of the continuous ranked probability score(CRPS) for isobaric variables in the Northern Hemisphere

    Fig. 5  Evolutions of the typhoon track forecast error and ensemble spread

    (a)averaged typhoon track forecast error and ensemble spread for all typhoon cases, (b)boxplot for the track forecast error, (c)boxplot for the track ensemble spread

    Fig. 6  Evolution of the area under relative operating characteristic curve(AROC) score for 24 h accumulated precipitation over China

    Fig. 7  Mean sea level pressure of the operational assimilation analysis field, mean sea level pressure of 72 h forecasts from the control forecast, the ensemble mean for ANSV and FCSV at 1200 UTC 31 Aug 2020(unit:hPa)

    (the isoline interval is 4 hPa)

    Fig. 8  Mean sea level pressure of 72 h forecasts from 5 ensemble members in ANSV experiment and FCSV experiment at 1200 UTC 31 Aug 2020(unit:hPa)

    (the isoline interval is 4 hPa)

    Table  1  Fraction of singular vector ensemble number based on similarity index between ANSV and FCSV at different intervals

    区域 奇异向量数量 奇异向量集合数量 相似指数在不同区间对应的奇异向量集合数量比例
    [0.0, 0.5) [0.5, 0.6) [0.6, 0.7) [0.7, 0.8) [0.8, 0.9) [0.9, 1.0]
    南北半球中高纬度目标区 30 38 0.000 0.000 0.526 0.474 0.000 0.000
    5 38 0.000 0.105 0.395 0.263 0.237 0.000
    热带气旋目标区 5 16 0.687 0.125 0.063 0.125 0.00 0.00
    注:南北半球中高纬度目标区指30°~80°N和30°~80°S地区,奇异集合数量为个例数量的2倍;热带气旋目标区指以热带气旋中心位置为中心的10°×10°区域,奇异向量集合数量为个例数量。
    DownLoad: Download CSV
  • [1]
    Leith C E. Theoretical skill of Monte Carlo forecasts. Mon Wea Rev, 1974, 102(6): 409-418. doi:  10.1175/1520-0493(1974)102<0409:TSOMCF>2.0.CO;2
    [2]
    Lorenz E N. A study of the predictability of a 28-variable atmospheric model. Tellus, 1965, 17(3): 321-333.
    [3]
    Mureau R, Molteni F, Palmer T N. Ensemble prediction using dynamically conditioned perturbations. Quart J Roy Meteor Soc, 1993, 119(510): 299-323.
    [4]
    Molteni F, Buizza R, Palmer T N, et al. The ECMWF ensemble prediction system: Methodology and validation. Quart J Roy Meteor Soc, 1996, 122(529): 73-119. doi:  10.1002/qj.49712252905
    [5]
    Buizza R. Potential forecast skill of ensemble prediction and spread and skill distributions of the ECMWF ensemble prediction system. Mon Wea Rev, 1997, 125(1): 99-119.
    [6]
    Leutbecher M, Palmer T N. Ensemble forecasting. J Comput Phys, 2008, 227(7): 3515-3539. doi:  10.1016/j.jcp.2007.02.014
    [7]
    Chen D H, Shen X S. Recent progress on GRAPES research and application. J Appl Meteor Sci, 2006, 17(6): 773-777. doi:  10.3969/j.issn.1001-7313.2006.06.014
    [8]
    Xue J S, Chen D H. Scientific Design and Application of Numerical Prediction System GRAPES. Beijing: Science Press, 2008.
    [9]
    Su Y, Shen X S, Zhang Q. Application of the correction algorithm to mass conservation in GRAPES_GFS. J Appl Meteor Sci, 2016, 27(6): 666-675. doi:  10.11898/1001-7313.20160603
    [10]
    Shen X S, Su Y, Hu J L, et al. Development and operation transformation of GRAPES global middle-range forecast system. J Appl Meteor Sci, 2017, 28(1): 1-10. doi:  10.11898/1001-7313.20170101
    [11]
    Liu Y, Xue J S. The new initialization scheme of the GRAPES. Acta Meteor Sinica, 2019, 77(2): 165-179. https://www.cnki.com.cn/Article/CJFDTOTAL-QXXB201902001.htm
    [12]
    Zhang L, Liu Y Z. The preconditioning of minimization algorithm in GRAPES global four-dimensional variational data assimilation system. J Appl Meteor Sci, 2017, 28(2): 168-176. doi:  10.11898/1001-7313.20170204
    [13]
    Liu Y Z, Zhang L, Jin Z Y. The optimization of GRAPES global tangent linear model and adjoint model. J Appl Meteor Sci, 2017, 28(1): 62-71. doi:  10.11898/1001-7313.20170106
    [14]
    Liu Y Z, Yang X S, Wang H Q. Research on GRAPES singular vectors and application to heavy rain ensemble prediction. Acta Sci Nat Univ Pekinensis, 2011, 47(2): 271-277. https://www.cnki.com.cn/Article/CJFDTOTAL-BJDZ201102014.htm
    [15]
    Liu Y Z, Shen X S, Li X L. Research on the singular vector perturbation of the GRAPES global model based on the total energy norm. Acta Meteor Sinica, 2013, 71(3): 517-526.
    [16]
    Li X L, Chen J, Liu Y Z, et al. Representations of initial uncertainty and model uncertainty of GRAPES global ensemble forecasting. Trans Atmos Sci, 2019, 42(3): 348-359. https://www.cnki.com.cn/Article/CJFDTOTAL-NJQX201903003.htm
    [17]
    Huo Z H, Li X L, Chen J, et al. The improved computation scheme for singular vectors based on hydrostatic equilibrium and application experiments using the GRAPES global model. Acta Meteor Sinica, 2021, 79(2): 282-299. https://www.cnki.com.cn/Article/CJFDTOTAL-QXXB202102008.htm
    [18]
    Li X L, Liu Y Z. The improvement of GRAPES global extratropical singular vectors and experimental study. Acta Meteor Sinica, 2019, 77(3): 552-562. https://www.cnki.com.cn/Article/CJFDTOTAL-QXXB201903013.htm
    [19]
    Huo Z H, Liu Y Z, Chen J, et al. The preliminary application of tropical cyclone targeted singular vectors in the GRAPES global ensemble forecasts. Acta Meteor Sinica, 2020, 78(1): 48-59. https://www.cnki.com.cn/Article/CJFDTOTAL-QXXB202001004.htm
    [20]
    Yuan Y, Li X L, Chen J, et al. Stochastic parameterization toward model uncertainty for the GRAPES mesoscale ensemble prediction system. Meteor Mon, 2016, 42(10): 1161-1175. https://www.cnki.com.cn/Article/CJFDTOTAL-QXXX201610001.htm
    [21]
    Peng F, Li X L, Chen J, et al. A stochastic kinetic energy backscatter scheme for model perturbations in the GRAPES global ensemble prediction system. Acta Meteor Sinica, 2019, 77(2): 180-195.
    [22]
    Lei Y, Guo Q Y, Qian Y, et al. Evaluation and quality mark of radiosonde geopotential height of L-band radar. J Appl Meteor Sci, 2018, 29(6): 710-723. doi:  10.11898/1001-7313.20180607
    [23]
    Hao M, Gong J D, Tian W H, et al. Deviation correction and assimilation experiment on L-band radiosonde humidity data. J Appl Meteor Sci, 2018, 29(5): 559-570. doi:  10.11898/1001-7313.20180505
    [24]
    Leutbecher M. On Ensemble prediction using singular vectors started from forecasts. Mon Wea Rev, 2005, 133(10): 3038-3046.
    [25]
    Huang Y Y, Xue J S, Wan Q L, et al. Improvement of the surface pressure operator in GRAPES and its application in precipitation forecasting in South China. Adv Atmos Sci, 2013, 30(2): 354-366.
    [26]
    Huang B, Chen D H, Li X L, et al. Improvement of the semi-Lagrangian advection scheme in the GRAPES model: The oretical analysis and idealized tests. Adv Atmos Sci, 2014, 31(3): 693-704.
    [27]
    Zhang M, Yu H P, Huang J P, et al. Assessment on unsystematic errors of GRAPES_GFS 2. 0. J Appl Meteor Sci, 2019, 30(3): 332-344. doi:  10.11898/1001-7313.20190307
    [28]
    Zhang M, Yu H P, Huang J P, et al. Assessment on systematic errors of GRAPES_GFS 2. 0. J Appl Meteor Sci, 2018, 29(5): 571-583. doi:  10.11898/1001-7313.20180506
    [29]
    Wang J, Wang B, Liu J J, et al. Application and characteristic analysis of the moist singular vector in GRAPES-GEPS. Adv Atmos Sci, 2020, 37(11): 1164-1178.
    [30]
    Wang J, Liu J J, Wang B, et al. A sensitivity study of the moist singular vectors to temporal and spatial scales in GRAPES-GEPS global ensemble prediction system. Chinese J Atmos Sci, 2021, 45(4): 874-888. https://www.cnki.com.cn/Article/CJFDTOTAL-DQXK202104012.htm
    [31]
    Simon H D. The Lanczos algorithm with partial reorthogonalization. Math Comput, 1984, 42(165): 115-142.
    [32]
    Buizza R. Sensitivity of optimal unstable structures. Quart J Roy Meteor Soc, 1994, 120: 429-451.
  • 加载中
  • -->

Catalog

    Figures(8)  / Tables(1)

    Article views (239) PDF downloads(42) Cited by()
    Proportional Views

    /

    DownLoad:  Full-Size Img  PowerPoint